Dissertation Defense of CMSE Spencer Lee
Department of Computational Mathematics, Science & Engineering
Michigan State University
Comprehensive Exam Notice
Friday, October 11th at 3:00-4:00 PM
EGR 1502 (Large CMSE Conference Room)
Zoom link: https://msu.zoom.us/j/98946723285
Passcode: 095570
Hermite-Based Discrete Adjoint Method for
Gradient Computation in Quantum Optimal Control
Spencer Lee
Abstract:
We present a new method for controlling quantum systems which utilizes an arbitrarily high-order method for solving systems of ODEs based on Hermite interpolation and the discrete adjoint method for efficiently computing derivatives of functions of finite sequences. The resulting quantum optimal control method is the first of its kind to use a high-order (higher than second order) timestepping method while computing a mathematically exact gradient with a computational complexity that is largely independent of the number of parameters needed to characterize the control of the quantum system. We compare the performance of our method, which is implemented in the open-source QuantumGateDesign.jl software package, with the state-of-the-art Juqbox.jl software package. We find that QuantumGateDesign.jl is able to simulate quantum systems to experimentally relevant levels of accuracy in a much shorter time than Juqbox.jl
Committee:
Andrew Christlieb (Co-chair)
Daniel Appelo (Co-chair)
Ryan LaRose
Johannes Pollanen
Dean Lee
